Re: Using Dismal V1.5 after updates

[Stefan Monnier <monnier@iro.umontreal.ca>]

I pushed a new version 1.5.2 to elpa.git which should appear on
elpa.gnu.org soon.  This should hopefully fix your problems (at least
the nb 2 and maybe nb 1 as well).


        Stefan


OURY, JACOB DAVID [2021-05-20 20:16:24] wrote:

> Hi, 
>
> My name is Jacob Oury and I'm a doctoral candidate advised by Frank Ritter (an original Dismal dev!)
> and I'm trying to get Dismal working on the latest version of EMACS. I've found a few issues, and I'm
> hoping you could provide some advice. I'll describe them below. 
>
> 1 I believe Dismal V1.5 is missing a key file for running dismal. Per the description, the package should
>  include a file called dismal-mode-defaults.el, however this is not available in the elpa.gnu package
>  repo. I was able to work around this issue by acquiring a previous version of Dismal from my
>  advisor and manually loading that file. Without that, I am unsure if I could get this package running.
>  Per the Git records at http://git.savannah.gnu.org/cgit/emacs/elpa.git/log/?h=externals/dismal, it
>  looks like that file was deleted in update below. I think this may have broken the package and I'm
>  hoping to find out why that file was deleted.  
>
>  author Stefan Monnier <monnier@iro.umontreal.ca> 2013-06-06 16:20:37 -0400 
>  committer Stefan Monnier <monnier@iro.umontreal.ca> 2013-06-06 16:20:37 -0400 
>  commit 554819b2d3c2e3d4494845a23deffd7245349a56 (patch) 
>  tree c4b11570454318e4dba291c1732f18e8dc639a58 
>  parent 18a2365ebc31e03bce64aa45227a5e6f5e869bad (diff) 
>  download elpa-554819b2d3c2e3d4494845a23deffd7245349a56.tar.gz 
>  General cleanup for GNU ELPA
> 2 Once I got the defaults loaded, I also am running into issues during testing (dis-test-dismal). It
>  initially runs, however, the test fails when trying to insert a simple math equation into a cell: 
>
>  dis-edit-cell-plain((+ 2 3))
>
>  While running the test, the process fails with the message:  "dismal-set-cell-internals: Symbol’s
>  value as variable is void: index". I believe this may be due to some effect of the outdated heaps.el
>  package, however Frank's version of Dismal (Dismal V1.4 running on Emacs 25.X, on MacOS) runs
>  fine. I'm using a Windows computer so that could play a role. When tracing the test through the
>  debugger, it fails immediately following a call to heap-empty with the info shown below. I'm still
>  figuring out the debugger, and I don't have a more thorough printout of the trace at the moment. 
>
>  Debugger entered--entering a function:
>  * command-error-default-function((void-variable index) "" dismal-set-cell-internals)
>    recursive-edit()
>  ;;Debugger cleared most of its text once the void variable error occurs, so the excerpts from
>  immediately before the error are below this line and the output from the debugger is above. 
>  * heap-empty([dismal-address-compare 1 0 [nil]])
>  * heap-member((3 . 0) [dismal-address-compare 1 0 [nil]])
>  * dismal-invalidate-cell((3 . 0))
>  * dismal-set-cell-internals(3 0 (+ 2 3) nil)
>  * dismal-set-cell(3 0 (+ 2 3) nil)
>  * dismal-edit-cell((+ 2 3) nil)
>
> I'm an EMACS novice (I generally work on the experimental cognitive psychology side of things while
> my labmates are more focused on using ACT-R for cognitive models), and so I may have made some
> missteps here, but I've been working with Frank during some debugging and still not having any luck
> so I'd appreciate any and all help you could provide on these issues. Thanks so much!
>
> Best regards, 
>
> Jacob D. Oury
> Doctoral Candidate in Informatics
> Applied Cognitive Science Lab
> Penn State College of Information Sciences and Technology